MC Merge browser: 'Accept same source' button considered harmful

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

MC Merge browser: 'Accept same source' button considered harmful

Chris Muller-3
Does anyone like to go around and change the format of other peoples
code and nothing else?  I consider this inconsiderate to my peers and
I'm sure most around here do, as well.

So does anyone know why in the world we this crazy button tacked onto
the end of the MC Merge browser?  Every time I have a merge conflict,
this button rears its ugly head and forces me to reckon with it.  Not
only is it incongruous with the rest of the buttons, but totallly
wrongheaded for how we respect each others work, resulting in
unpleasant thoughts and experience every time I see it.

Browsing origin just shows:
___
     Name: Monticello-tfel.637
     Author: tfel
     Time: 25 July 2016, 3:24:24.996828 pm
     UUID: cf0d6af1-c703-5044-9c57-e798b0cf3abf
     Ancestors: Monticello-cmm.636

     add a button for rejecting all incoming conflicts that only change AST
____

(rant) Explaining WHAT your changed in your version notes is
redundant, we can all see the code.  Please tell us WHY you are doing
it.

So does anyone know?  I think we should remove it and I cringe to
wonder how many original works we lost to it...   :(

Reply | Threaded
Open this post in threaded view
|

Re: MC Merge browser: 'Accept same source' button considered harmful

David T. Lewis
On Sat, Jun 23, 2018 at 06:39:51PM -0500, Chris Muller wrote:

> Does anyone like to go around and change the format of other peoples
> code and nothing else?  I consider this inconsiderate to my peers and
> I'm sure most around here do, as well.
>
> So does anyone know why in the world we this crazy button tacked onto
> the end of the MC Merge browser?  Every time I have a merge conflict,
> this button rears its ugly head and forces me to reckon with it.  Not
> only is it incongruous with the rest of the buttons, but totallly
> wrongheaded for how we respect each others work, resulting in
> unpleasant thoughts and experience every time I see it.
>
> Browsing origin just shows:
> ___
>      Name: Monticello-tfel.637
>      Author: tfel
>      Time: 25 July 2016, 3:24:24.996828 pm
>      UUID: cf0d6af1-c703-5044-9c57-e798b0cf3abf
>      Ancestors: Monticello-cmm.636
>
>      add a button for rejecting all incoming conflicts that only change AST
> ____
>
> (rant) Explaining WHAT your changed in your version notes is
> redundant, we can all see the code.  Please tell us WHY you are doing
> it.
>
> So does anyone know?  I think we should remove it and I cringe to
> wonder how many original works we lost to it...   :(
>

I do not understand the rant. The offending button was added almost a year
ago, and its balloon help says this:

  Choose all local conflicting versions that have essentially the same code

Even though I have never used this, the intent seems clear. It says "if
somebody was just messing around with the formatting but did not change
anything, then reject the change because it is not nice to mess with the
format of other peoples' code."

What's not to like about that?

Dave
 

Reply | Threaded
Open this post in threaded view
|

Re: MC Merge browser: 'Accept same source' button considered harmful

Tobias Pape

> On 24.06.2018, at 02:57, David T. Lewis <[hidden email]> wrote:
>
> On Sat, Jun 23, 2018 at 06:39:51PM -0500, Chris Muller wrote:
>> Does anyone like to go around and change the format of other peoples
>> code and nothing else?  I consider this inconsiderate to my peers and
>> I'm sure most around here do, as well.
>>
>> So does anyone know why in the world we this crazy button tacked onto
>> the end of the MC Merge browser?  Every time I have a merge conflict,
>> this button rears its ugly head and forces me to reckon with it.  Not
>> only is it incongruous with the rest of the buttons, but totallly
>> wrongheaded for how we respect each others work, resulting in
>> unpleasant thoughts and experience every time I see it.
>>
>> Browsing origin just shows:
>> ___
>>     Name: Monticello-tfel.637
>>     Author: tfel
>>     Time: 25 July 2016, 3:24:24.996828 pm
>>     UUID: cf0d6af1-c703-5044-9c57-e798b0cf3abf
>>     Ancestors: Monticello-cmm.636
>>
>>     add a button for rejecting all incoming conflicts that only change AST
>> ____
>>
>> (rant) Explaining WHAT your changed in your version notes is
>> redundant, we can all see the code.  Please tell us WHY you are doing
>> it.
>>
>> So does anyone know?  I think we should remove it and I cringe to
>> wonder how many original works we lost to it...   :(
>>
>
> I do not understand the rant. The offending button was added almost a year
> ago, and its balloon help says this:
>
>  Choose all local conflicting versions that have essentially the same code
>
> Even though I have never used this, the intent seems clear. It says "if
> somebody was just messing around with the formatting but did not change
> anything, then reject the change because it is not nice to mess with the
> format of other peoples' code."

+10

>
> What's not to like about that?
>
> Dave
>
>


Reply | Threaded
Open this post in threaded view
|

Re: MC Merge browser: 'Accept same source' button considered harmful

Chris Muller-3
In reply to this post by David T. Lewis
I now understand the scenario you're thinking of -- Etoys had a lot of
same-source-different-timestamps methods, and they wanted to preserve
the timestamps in Squeak.  That's fine, except it's possible Etoys
might have had the earlier version in some cases -- I'm thinking about
how many "sd 2006"'s we may still have...

The scenario I was thinking of, the _local_ version is the reformatted
one, while the one being merged is the original.  For example, someone
puts a "self halt" in a method, then removes it instead of reverts the
method.  Then, they update from trunk and get a conflict on that
method.  In that scenario, that button is doing exactly the wrong
thing.

The more appropriate place to handle this issue is on the Save MC
Version panel (because we self review every change of every commit,
right?   :)  ).  That's where we already see the "same source
different timestamps" and can easily revert them right there with
Command+x.  Only the reviewer has sufficient context to know which
version to to choose, this button cannot possibly relieve one of that
responsibility.  If no one objects, I'd like to remove it.

 - Chris
On Sat, Jun 23, 2018 at 7:57 PM David T. Lewis <[hidden email]> wrote:

>
> On Sat, Jun 23, 2018 at 06:39:51PM -0500, Chris Muller wrote:
> > Does anyone like to go around and change the format of other peoples
> > code and nothing else?  I consider this inconsiderate to my peers and
> > I'm sure most around here do, as well.
> >
> > So does anyone know why in the world we this crazy button tacked onto
> > the end of the MC Merge browser?  Every time I have a merge conflict,
> > this button rears its ugly head and forces me to reckon with it.  Not
> > only is it incongruous with the rest of the buttons, but totallly
> > wrongheaded for how we respect each others work, resulting in
> > unpleasant thoughts and experience every time I see it.
> >
> > Browsing origin just shows:
> > ___
> >      Name: Monticello-tfel.637
> >      Author: tfel
> >      Time: 25 July 2016, 3:24:24.996828 pm
> >      UUID: cf0d6af1-c703-5044-9c57-e798b0cf3abf
> >      Ancestors: Monticello-cmm.636
> >
> >      add a button for rejecting all incoming conflicts that only change AST
> > ____
> >
> > (rant) Explaining WHAT your changed in your version notes is
> > redundant, we can all see the code.  Please tell us WHY you are doing
> > it.
> >
> > So does anyone know?  I think we should remove it and I cringe to
> > wonder how many original works we lost to it...   :(
> >
>
> I do not understand the rant. The offending button was added almost a year
> ago, and its balloon help says this:
>
>   Choose all local conflicting versions that have essentially the same code
>
> Even though I have never used this, the intent seems clear. It says "if
> somebody was just messing around with the formatting but did not change
> anything, then reject the change because it is not nice to mess with the
> format of other peoples' code."
>
> What's not to like about that?
>
> Dave
>
>

Reply | Threaded
Open this post in threaded view
|

Re: MC Merge browser: 'Accept same source' button considered harmful

Tobias Pape

> On 24.06.2018, at 19:42, Chris Muller <[hidden email]> wrote:
>
> I now understand the scenario you're thinking of -- Etoys had a lot of
> same-source-different-timestamps methods, and they wanted to preserve
> the timestamps in Squeak.  That's fine, except it's possible Etoys
> might have had the earlier version in some cases -- I'm thinking about
> how many "sd 2006"'s we may still have...
>
> The scenario I was thinking of, the _local_ version is the reformatted
> one, while the one being merged is the original.  For example, someone
> puts a "self halt" in a method, then removes it instead of reverts the
> method.  Then, they update from trunk and get a conflict on that
> method.  In that scenario, that button is doing exactly the wrong
> thing.
>
> The more appropriate place to handle this issue is on the Save MC
> Version panel (because we self review every change of every commit,
> right?   :)  ).  That's where we already see the "same source
> different timestamps" and can easily revert them right there with
> Command+x.  Only the reviewer has sufficient context to know which
> version to to choose, this button cannot possibly relieve one of that
> responsibility.  If no one objects, I'd like to remove it.

Remove what? the button in the merge panel?
I object :)

Best regards
        -Tobias

>
> - Chris
> On Sat, Jun 23, 2018 at 7:57 PM David T. Lewis <[hidden email]> wrote:
>>
>> On Sat, Jun 23, 2018 at 06:39:51PM -0500, Chris Muller wrote:
>>> Does anyone like to go around and change the format of other peoples
>>> code and nothing else?  I consider this inconsiderate to my peers and
>>> I'm sure most around here do, as well.
>>>
>>> So does anyone know why in the world we this crazy button tacked onto
>>> the end of the MC Merge browser?  Every time I have a merge conflict,
>>> this button rears its ugly head and forces me to reckon with it.  Not
>>> only is it incongruous with the rest of the buttons, but totallly
>>> wrongheaded for how we respect each others work, resulting in
>>> unpleasant thoughts and experience every time I see it.
>>>
>>> Browsing origin just shows:
>>> ___
>>>     Name: Monticello-tfel.637
>>>     Author: tfel
>>>     Time: 25 July 2016, 3:24:24.996828 pm
>>>     UUID: cf0d6af1-c703-5044-9c57-e798b0cf3abf
>>>     Ancestors: Monticello-cmm.636
>>>
>>>     add a button for rejecting all incoming conflicts that only change AST
>>> ____
>>>
>>> (rant) Explaining WHAT your changed in your version notes is
>>> redundant, we can all see the code.  Please tell us WHY you are doing
>>> it.
>>>
>>> So does anyone know?  I think we should remove it and I cringe to
>>> wonder how many original works we lost to it...   :(
>>>
>>
>> I do not understand the rant. The offending button was added almost a year
>> ago, and its balloon help says this:
>>
>>  Choose all local conflicting versions that have essentially the same code
>>
>> Even though I have never used this, the intent seems clear. It says "if
>> somebody was just messing around with the formatting but did not change
>> anything, then reject the change because it is not nice to mess with the
>> format of other peoples' code."
>>
>> What's not to like about that?
>>
>> Dave
>>
>>
>